W&B Deep Learning Salon - Reproducibility Challenge

W&B’s Deep Learning Salon, featuring talks and Q&As with researchers, practitioners, hackers, and more in the Deep Learning space. This salon features W&B community members who worked in teams to reproduce ML research papers from last year’s top conferences. Learn more about how to take part in the next W&B Reproducibility Challenge, and get a $500 subsidy, here: Read the Reports: - Reproducing RigL, by Rajat Dwaraknath and Varun Sundar: - Reproducing Reformer, by Morgan McGuire, Hallvar Gisnas, Arto Arutiunian: Video Contents: 0:00 The Reproducibility Challenge 3:25 Reproducing Reformer, by the community team 19:40 Discussion of Reformer project results and workflow 33:40 Reproducing RigL, by Varun Sundar and